UPDATED 09:30 EDT / MAY 11 2014

Testing all-flash arrays made easy, courtesy of EMC XtremIO  | #EMCworld

Miroslav Klivanski - EMC World 2014 - theCUBEAll Flash Arrays offer a number of distinct advantages over traditional disk systems, starting with superior performance and lower latency, but assessing the impact of that increased responsiveness on the bottom line involves a different set of metrics that many IT professionals in the traditional enterprise are not necessarily comfortable with yet.

That knowledge divide can turn into a major problem because without a clear view of the benefits,  an organization has no way of accurately determining whether or not an all-SSD system such as the XtremIO Array is worth the premium price tag, which hampers adoption and hurts sales for vendors. EMC, however, is not taking this lying down.  Miroslav Klivansky, the chief technologist for the vendor’s XtremIO unit,  stopped by theCUBE during its recently concluded customer conference in Las Vegas to share what his team is doing to make easier for CIOs to justify their solid-state memory investments with siliconANGLE founder John Furrier and Wikibon chief analyst Dave Vellante.

“Doing apples-to-apples comparisons for storage arrays is always a little tricky,” and it becomes even more difficult when it comes to measuring the cost-performance ratio of a flash system against that of a another or a traditional disk solution, Klivansky highlights. The challenge lies in the method of evaluation.

Flash is a “completely different beast” from a technical standpoint, he explains, necessitating a unique approach to performance testing and specialized tools. To that end, XtremIO has created a purpose-built assessment suite based on vdbench, a popular command line utility for validating storage systems that was originally developed by Sun Microsystems, now a part of Oracle.

Klivansky says that one of the reasons his team chose the tool is that it’s open source and, as a result, vendor-neutral, which means than it can be used to compare a wide range of systems from different suppliers. The software was also a very attractive candidate in terms of the functionality, he continues: “It’s very savvy around generating data that can work with storage efficiency algorithms, whether you need to dial in a certain dedupe or compression factor or you want to avoid that.”

The suite implements IDC’s official framework for evaluating all-flash arrays, according to Klivansky. The first step is preconditioning,  or writing to every single cell in a system before doing any testing, a process that he explains is made necessary by the fact that unlike disk, the performance of solid-state drivers degrades over time.

Once an all-flash array has been properly preconditioned, the actual evaluation can begin.  Klivansky says an array has to be testing over a broad spectrum of workloads with different read/write ratios for accurate results.

“Flash is fast. You can throw a bunch of flash into a box, and there are solutions out there that do, but just because the flash is fast doesn’t mean it’s gonna be fast in all different conditions, under all different circumstances, under heavy load as well as light load,” he elaborates. ”So if all you’re doing is throwing a couple of VMs on there and kinda lightly tickling the box, you’re not gonna get a real impression of what this array can do.”

The XtremIO version of vdbench was build with these requirements in mind. “It’s meant up to be set up and run over 2-3 days of heavy, I/O-intensive load on the array and it will characterize a lot of different combinations and give you information about how it thinks it will perform under stress in an enterprise environment,”  Klivansky notes. He adds that the analytics capabilities of the the toolkit are complemented by built-in data visualization functionality allowing admins to create graphs and charts which make performance statistics more accessible.


Since you’re here …

… We’d like to tell you about our mission and how you can help us fulfill it. SiliconANGLE Media Inc.’s business model is based on the intrinsic value of the content, not advertising. Unlike many online publications, we don’t have a paywall or run banner advertising, because we want to keep our journalism open, without influence or the need to chase traffic.The journalism, reporting and commentary on SiliconANGLE — along with live, unscripted video from our Silicon Valley studio and globe-trotting video teams at theCUBE — take a lot of hard work, time and money. Keeping the quality high requires the support of sponsors who are aligned with our vision of ad-free journalism content.

If you like the reporting, video interviews and other ad-free content here, please take a moment to check out a sample of the video content supported by our sponsors, tweet your support, and keep coming back to SiliconANGLE.